3D Reconstruction Of Maxillary Central Incisors Using Deep Learning Algorithms

This study aims to develop a deep learning-based three-dimensional point cloud reconstruction network that utilizes the morphological features of the ipsilateral maxillary lateral incisor, canine, and premolars to predict and reconstruct the anatomical morphology of the maxillary central incisor, providing insights for personalized anterior tooth aesthetic restoration. Exocad 3.2 was used to segment crowns of teeth #11-14. After standardizing coordinate systems in MATLAB 2019a, 182 cases were randomly selected as the training set and 10 as the test set. A dual-network architecture (morphology and pose estimation) was trained to reconstruct maxillary central incisor point clouds. Reconstructed outputs were meshed, aligned with original crowns for evaluation, and applied to incisal third defect models of tooth #11 to compare restoration outcomes among the proposed algorithm, mirroring technique, and standard database method. The test set achieved a Chamfer Distance of 0.405, Earth Mover’s Distance of 0.152, and root mean square error of 0.128 ± 0.030 mm. For incisal defect restoration, the proposed algorithm (0.128 ± 0.030 mm) demonstrated comparable accuracy to mirroring (0.130 ± 0.021 mm, p = 0.865) but significantly outperformed the database method (0.233 ± 0.038 mm, p < 0.001). The proposed 3D point cloud reconstruction algorithm enables high-precision maxillary central incisor restoration based on adjacent tooth morphology, offering a reliable technical solution for personalized anterior dental rehabilitation.